CHAPTER I1

REVIEW OF LITERATURE

This chapter reviews various empirical studies related to the linkages

between the stock prices and the exchange rates. This review has been divided

into two sections. The first section deals with studies related to empirical

validity of market integration hypothesis with reference to foreign exchange

market and stock market and inter-relationship between the two at the

international level. The second classification deals with studies pertaining to

the validity of market integration hypothesis in foreign exchange market and

stock market and inter-relationship between the two at the national level. This

classification is done to examine the various empirical issues of the relation

between foreign exchange rates and stock prices. This classification again

helps us to find the studies which are giving conflicting results, and also to

identifying the gap with special reference to India. Besides, the studies which

are giving conflicting results and have not received due importance should be

considered for further empirical research. With these objectives in mind the

empirical studies have been discussed below:

International-level Studies

Eur and Resnick (1988) tried to develop ex ante portfolio selection

strategies to realize potential gains from international diversification under

flexible exchange rates. For the empirical analysis the Morgan Stanley Capital

lntemational Perspective daily stock index values for the United States and the

other six countries were adopted. The stock indices of United States, Canada,

France, Germany, Japan, Switzerland, and the U.K. were value weighted and

it was a representative of a domestic stock index fund. The data series were

provided in both the United States and the local currencies for the period from

December 3 1, 1979, through December 10, 1985. Methods such as correlation,

variance and covariance have also been employed to know the changes in

stock market across the countries

The analysis reveals that exchange rate uncertainty is a largely non-

diversifiable factor adversely affecting the performance of international

portfolios. The authors have suggested two methods such as multi-currency

diversification and hedging via forward exchange contracts for reducing the

exchange rate risks.

Ma and Kao (1990) examined the stock price reactions to the

exchange rate changes. The authors have studied the case of six developed

countries namely United Kingdom, Canada, France, West Germany, Italy and

Japan. Monthly stock indices and monthly exchange rates arc gathered from

the Exchange Rates and Interest Rates Tape Provided by the Federal Reserve.

The sample period was from January 1973 to December 1983, and a two

factor model was adopted for the empirical analysis.

The paper demonstrates two possible impacts of changes in a country's

currency value on stock price movements. Firstly, the financial effects of

exchange rate changes on the transaction exposure. Secondly, the economic

effect from exchange rate changes suggests that, for an export-dominant

country, the currency appreciation reduces the competitiveness of export

markets and has a negative effect on the domestic stock market. On the other

hand for an import dominated country, the currency appreciation will lower

import costs and generate a positive impact on the stock market.

Jorion (1991) examined the pricing of exchange rate risk in the United

States (US) stock market, by using two-factor and multi-factor arbitrage

pricing models. For the purpose of empirical analysis, monthly data are

collected for a period ranging from January 1971 to December 1987. The data

on the trade-weighted exchange rate is derived from the weights in the

Multilateral Exchange Rata Model (MERM) computed by the International

Monetary Fund (IMF). Monthly data on the Stock market return are collected

from the University of Chicago's Centre for Research in Security Prices

(CRSP) database. An ordinary least squares (OLS) regression method was

employed for examining the objective.

The study reveals that United States (US) industries display significant

cross-sectional differences in their exposure to movements in the dollar.

However, the empirical results do not suggest that exchange rate risk is priced

in the stock market. The conditional risk premium attached to foreign

currency exposure appears to be small and never significant.

Bartov and Bodnar (1994) re-examined the anticipated changes in the

dollar and equity value. The period of study ranges from the fiscal year 1978

and runs through the fiscal year 1989. The authors have used the

COMPUSTAT Merged-Expanded Annual Industrial File and Full Coverage

File for fums that reported significant foreign currency gains or losses on their

annual financial statements. The data on stock prices were collected either the

Centre for Research in Security Prices (CRSP) New York Stock Exchange

(NYSE)lAmerican Stock Exchange (AMEX) Daily Return File or the National

Association of Security Dealers Automated Quotation (NASDAQ) Daily or

Master Files.

The results of the study show that contemporaneous changes in the

dollar have little power in explaining abnormal stock return. This finding is

consistent with the failure of prior research to document a contemporaneous

relation between dollar fluctuations and fum value and suggests that problems

with sample selection technique are not a complete explanation for their

failure.

Choi and Prasad (1995) estimated a model of firm valuation to examine

the exchange risk sensitivity of firm value. For the empirical analysis monthly

time-series of stock returns were obtained from the University of Chicago

Centre for Research in Security Prices (CRSP) tapes and COMPUSTAT

htabase. The period of study was from January 1978 to December 1989. The

nominal exchange rate variable was the United States (US) dollar value of one

unit of foreign currency, where foreign cumncy was the multilateral trade-

weighted basket of ten major currencies as published in the Federal Reserve

Bulletin. The o r d i i least squares (OLS) and the generalized least squares

(GLS) methods were employed for examining the objective of the study.

The study reveals that approximately 60 per cent of the firms with

significant exchange risk exposure benefited, and 40 per cent lost, with a

depreciation of the dollar. It is also found that the exchange rate factor is less

important in explaining the industry portfolios.

Prasad and Rajan (1995) investigated the impact of exchange rate

fluctuation on equity valuation in Germany, Japan, the United Kingdom and

the United States. They also evaluated the existence of exchange risk premia

in these four markets. For the empirical analysis, monthly data on United

States (US) stocks were collected from the Centre for Research in Security

Prices (CRSP) tapes and the data for other countries came from the Morgan

Stanley Capital International Perspective for the period from January 1981 to

December 1989. The exchange rate variable is defined as the home currency

value of one unit of foreign currency, where foreign currency is defined as a

trade - weighted basket of 16 currencies. The basket of currencies was derived

using 1981 trade weights computed by the International Monetary Fund using

the Multilateral Exchange Rate Model (MERM). The exchange risk exposure

coefficients are estimated using the Seemingly Unrelated Regression (SUR)

procedure. The pricing coefficients were estimated jointly with the sensitivity

coefficients using the iterated non linear seemingly unrelated (ITNLSUR)

procedure.

The study finds evidence of exchange rate risk sensitivity in each of the

four markets to varying degrees, with the German and the United States (US)

markets yielding a maximum number of industries with significant exchange

rate exposure. Export-oriented industries gaining from a depreciation of the

home currency and import-dominated industries gaining from an appreciation

for the home currency.

Bartov et a1 (1996) attempted to reveal the relation between exchange

rate variability and stock return volatility for United States (US) multinational

f m . Tests cover two five-year periods i.e., five years of fixed rates prior to

the break-down of the Bretton Woods system (1966-1970) and first five years

following the anival of fluctuating exchange rates (1973-1977). For the

empirical analysis, a sample of 109 firms with foreign operations was

identified. Monthly return data of all firms were collected from the Centre

for Research in Security Prices (CRSP) New York Stock Exchange

OJYSE) or American Stock Exchange (AMEX), Monthly Returns and Master

File. The exchange rate index is a weighted index of the United States dollar

against the other G-7 countries, and it was collected from the Multilateral

Exchange Rate Model (MERM) of the International Monetary Fund (IMF).

Chi-squared test statistic was employed to evaluate the significance of the

change in stock return variance for each sample across the two periods.

Further, regression technique has been adopted for the purpose of estimating

monthly return to f m .

The analysis confirms that the increase in total stock return volatility

for multinational finns is significantly larger than that of other firms. The

result also supports the weak existing empirical evidence that exchange rates

play an important role in the stock price behaviour of United States (US)

multinational firms.

Chamberlain et a1 (1997) examined the foreign exchange exposure of a

sample of United States (US) and Japanese banking firms. In constructing the

United States (US) sample, both daily and monthly stock returns of thirty bank

holding companies that were traded over the entire sample period from 1986

to 1992 on the New York Stock Exchange (NYSE) or the American Stock

Exchange (AMEX) were selected from the Centre for Research in Security

Prices (CRSP). For Japanese bank samples, monthly observations of the

largest 110 Japanese bank returns were collected from Worldscope data, and

daily bank retwns were considered from extel Research data. The authors

estimated the sensitivity of returns to the exchange rate in the context of an

augmented market model.

The analysis reveals that the stock returns of approximately one-third of

thirty large United States (US) bank holding companies appear to be sensitive

to exchange rate changes. Whereas few Japanese bank return, appear to be

sensitive to exchange rate changes. The authors leave the causes for the

difference between the exchange rate sensitivities of Japanese and United

States (US) banking firms to the h r e research.

Mookcjee and Yu (1997) tested for the presence of informational

inefficiencies in the Singapore stock market using a subset of macroeconomic

variables, including nominal exchange rates, which are pertinent in the context

of a small open economy. All monthly data used in the study were from

October 1984 through April 1993. Tests of nonstationarity have been

conducted using both the Augmented Dickey Fuller (ADF) and KPSS test

statistics. The authors employed the techniques of cointegration, causality and

forecasting equations to explore the nexus between Singapore stock returns

and its four macro variables.

The study found that exchange rates do not exhibit a long run

equilibrium relationship with stock prices. Tests of causality and forecasting

equations yield different results. While the causality test approach has

revealed market efficiency with respect to exchange rates, the forecasting

equation approach has demonstrated market inefficiency. Hence, the analysis

concluded that the results based on only one or the other approach should be

viewed with caution.

A ~ ~ O N O U ~ et al (1998) investigated the behaviour of the equity market

risk premium for the London Stock Exchange prior to and during Sterling

membership of the Exchange Rate Mechanism (ERM). The authors have used

monthly data from January 1980 to August 1993 for both the macroeconomic

variables and the individual security returns on the 69 companies traded on the

London Stock Exchange. All necessary information, including the Pound

SterlingDeutsch Mark (£DM) exchange rate were collected from the

Datastream database. The Arbitrage Pricing Theory (APT) was used as the

model of the equity risk premium for empirical analysis.

The authors observed that the membership of the Exchange Rate

Mechanism (ERM) and the resultant increased convergence between Britain

and the other member states were beneficial to the United Kingdom with a

reduction in both the exchange rate risk premium and the total equity market

risk premium. The study conveys the importance of a policy of convergence

and exchange rate stability maintained through a fixed exchange rate regime in

terms of a reduction in uncertainty. Experience of United Kingdom prior to

and during sterling membership of the European Exchange Rate Mechanism

(ERM) clearly reveals the above idea.

Malliaropulos (I 998) investigated the link between international stock

return differentials relative to the United States and deviation from relative

Purchasing Power Parity. The end-of-quarter stock indices for the G5

countries and nominal United States dollar exchange rate, from

DATASTREAM, and consumer price indices from the Organisation for

Economic Co-operation and Development (OECD) Quarterly National

Accounts database were collected for a period from the first quarter of 1973 to

the third quarter of 1992. In order to account for the endogeneity of the

regressor and small sample bias in the context of regression with overlapping

observations, the author has estimated sampling distribution of slope

coefficients and t-statistics using a distribution - free bootstrap technique

based on the Vector Auto Regressive (VAR) model.

The empirical results indicate that there is strong evidence of a negative

relationship between international stock return differential, and changes in the

real exchange rate for France and weaker evidence for Japan and United

Kingdom. The practical implication is that long-horizon return of United

States equity funds are likely to out perform comparable funds in other

countries when the dollar is overvalued relative to Purchasing Power Parity.

Morley and Pentecost (1998) enunciated the relationship between the

foreign exchange risk premium and excess returns in the national and foreign

equity markets in G-7 countries using both the United States dollar and the

United Kingdom pound as base currencies. All monthly time series data on

exchange rates for the period January 1982 to January 1994 were extracted

from Datastream database. Bilateral exchange rates are all measured against

the United States dollar and taken from International Financial Statistics.

Share prices of all the G-7 countries were taken from the corresponding stock

market indices for the same period. Analysis has been canied out by using the

cointegration methodology with an Error Correction Model (ECM) fimework

to capture the timeseries dynamics. Seemingly Unrelated Regression (SUR)

system estimation approach of Zellner (1962) was also used to improve the

efficiency of the estimated short-mn parameters.

The results show that in equilibrium, the foreign exchange market risk

premium is positively related to the domestic excepted equity premium and

negatively related to the foreign equity premium for all countries against both

exchange rates.

Vansconcellos and Kish (1998) analysed the influence of

macroeconomic variables, including the exchange rates on the number and

direction of cross-border acquisition between firms in the United States and

each of four European Countries, namely, Germany, Italy, the United

Kingdom and France. The data for the dependent variable and the explanatory

variables were obtained on a quarterly basis from 1982 through 1994. The

number of completed acquisitions was obtained from the several issues of

'Mergers and Acquisitions'. The data for exchange rates were downloaded

from the Datastream database. The logit model and the ordinary least squares

(OLS) regression are used to analyse select factors affecting cross-border

merger.

The study reveals the importance of exchange rate as a significant

explanatory variable of acquisitions for the fmt three countries except France.

A strong United States dollar may persuade firm to acquire a foreign firm

because the initial acquisition price is favourable when measured in dollar

terms in a net present value calculation. The authors conclude underlining the

role of exchange rate as a predictor of @ends in acquisition.

Wetmon and Brick (1998) made an attempt to estimate various risk

components of commercial bank stock returns. They argue that over time, the

relative significance of interest rate risk has given way to foreign exchange

risk and more recently to the basis risk. For the empirical analysis weekly

stock return data of a sample of 66 commercial banks for the period January 1,

1986 through June 30, 1995 were collected from Standard and Poor's Daily

Stock Price Record. Foreign exchange risk was proxied by the index of

weighted average exchange value of a dollar against ten currencies. For this

purpose monthly reported data published by the Federal Reserve Bulletin were

interpolated to generate weekly data. To get the explanatory power of different

variables including the exchange rate, an equation was estimated using

ordinary least square (OLS).

The results of the study appear quite interesting. All commercial banks

showed significant interest rate sensitivity before October 20, 1987. After

October 20, interest rate sensitivity gave way to foreign exchange risk

sensitivity. Between November 27, 1989 and January 7, 1991, foreign

exchange risk increased for all the banks. The authors concluded emphasising

the importance of significant basis risk component after June 10, 1994 for all

banks.

Bodart and Reding (1999) examined the behaviour of domestic daily

returns on bond and stock markets with the objective of identifying whether

there exist significant differences in the patterns of volatilities and

international correlations between European Exchange Rate Mechanism

(ERM) and non- European Exchange Rate Mechanism (ERM) countries. For

the empirical analysis countries were divided into two groups. The first group

is composed of hardcore European Monetary Systems (EMS) countries

(Germany, France and Belgium) which have maintained a fixed bilateral

exchange rate during the whole sample period; the other set includes 'drop-out

countries' (Italy, the Unitrd Kingdom and Sweeden) which abandoned the peg

with the Deutschmark (DEM) in fall 1992 and which have been on a float

thereafter. All necessary information came ffom Datastream, stock prices were

obtained from the six different indices. All prices and exchange rates are daily

closing prices and are expressed in domestic currency. Data were collected

over the period from January 2, 1989 to December 19, 1994. Estimates of

conditional variances were obtained by estimating over the whole sample

period a univariate Generalised Autoregressive Conditional Heteroskedasticity

(1,l) model for each series of daily returns. The extent to which changes in the

exchange rate regime affect the conditional international correlations among

asset markets also has been analysed.

The analysis failed to detect a marked linkage between the pattern of

volatilities on the stock market and the exchange market. However, the degree

of exchange rate variability exerts an influence on international bond and

stock correlations.

Ding et a1 (1999) used transaction data for the Kuala Lumpur Stock

Exchange (KLSE) and the Stock Exchange of Singapore (SES) for a major

Malaysian conglomerate, Sime Darby Berhad, and intraday exchange rate data

to examine whether to what extent each exchange contributes to price

discovery. The Stock Exchange of Singapore transaction &ta were down-

loaded from a United States commercial vendor. The Kuala Lumpur Stock

Exchange transaction data of Sime Darby Berhad were obtained for the period

from May 22, 1996 to July 17 1996. Data on Singapore dollar / Malaysian

ringgit exchange rates at four times during the trading day have been collected

from a brokerage firm. To test the presence of a unit root in the individual

series the augmented Dickey Fuller (1981) test was employed. The reduced

rank regression procedure of Johansen (1988) was also employed to test for

possible cointegration.

The empirical results reveal that the price series of stock exchange and

exchange rate for both Malaysian and Singapore are cointegrated. Arbihage

opportunities that appear to be present using unadjusted price series are

adjusted for exchange rates.

Doukas et a1 (1999) examined whether exchange rate risk is priced in

the equity market of Japan. Monthly time series of stock returns for 1079

Japanese firms traded on the Tokyo Stock Exchange for the period kom

January 1975 to December 1995 have been used in the study. Return data was

obtained from Worldscope, 1996, Datastream, and the Pacific Basin Capital

Market (PACAP) Database. The Stopford Dictionary of Multinations was used

to determine firms multinational status. The end-of-month exchange rate for

the Japanse Yen against the United States dollar (bilateral) and a real, moving,

trade-weighted exchange rate (multilateral) published by the Bank of England

were retrieved from Datastream. Both factors - sensitivities and risk premia -

were estimated using the Iterated Nonlinear Seemingly Unrelated Regression

Equation (INSURE) procedure.

The tests reveal that the foreign exchange risk premium is a significant

component of Japanse stock returns. The currency-risk exposure commands a

significant risk premium for multinationals and high - exporting Japanese

firms, whereas it is less in case of low - exporting and domestic firms.

Kwon and Shin (1999) examined the cointegration and causality

between macroeconomic variables, including the exchange rate, and stock

market returns in Korea. The data for the analysis were from monthly stock

prices of the value-weighted Korea composite Stock Prices lndex (KOSPI)

and small size stock price index (SMLS) for the period from January 1980 to

December 1992 (156 monthly observations), and these date information were

collected from various issues of the securities statistics yearbook. The stock

prices are the closing prices of the last trading day in each month. The macro

economic variables considered for the analysis are monthly data for the same

~eriod collected from various issues of the monthly Bulletin published by the

Bank of Korea. The Augumented Dickey Fuller (ADF) test was employed to

test the unit root hypothesis. The Vector El~or Correction Model (VECM) was

employed to test the presence of cointegration between the variables.

The analysis concluded that a set of variables are cointegrated with

stock market indexes, even though there is no cointegration between stock

prices indexes and any macroeconomic variables. The study also reveals that

the past changes in exchange rate significantly affect current changes in Korea

composite Stock Prices Index (KOSPI).

Friberg and Nydahl (1999) examined the exchange rate exposure of

national stock markets. The authors have investigated the relationship

between the valuation of the stock market and a trade weighted exchange rate

index for 11 industrialized countries. For the analysis, monthly data for the

period 1973-1996 were considered. The Morgan-Stanley stock market indexes

and world market index were from Morgan-Stanley. Data on nominal

exchange rates and local stock market data were collected from the Ecowin

database. The analysis is carried out using an ordinary least squares (OLS)

regression.

The study has established a positive relationship between stock market

exposure to exchange rates and the openness of a country for 11 industrialized

countries in the post Bretton-Woods period. It is also revealed that the more

open the economy, the stronger is the relationship between return on the stock

market and the exchange rate.

Amain and Hook (2000) investigated the relationship between the

exchange rate of Malaysian ringgit in t e n s of United States dollar and stock

prices in Kuala Lumpur Stock Exchange (KLES) using the single-index and

multi-index models. The authors have used 256 weekly closing stock price

indices and the Malaysian RinggitrUnited States Dollar (RM/US$) exchange

rate spanning from September 1993 to July 1998. The data were collected

from the Daily Diary published by Kuala Lumpur Stock Exchange (KLSE).

The study period was divided as a cycle of a strong ringgit (September, 93-

January, 97) and a cycle covering a weak ringgit (July, 97-July, 98). Besides

an ordinary least square (OLS) method was cmployed to identify the

relationship between stock prices and exchange rate.

The results of the study indicate no significant relationship between the

exchange rate and the Kuala Lumpur Stock Exchange (KLSE) stocks during

strong ringgit period. But for the weak ringgit period a weak relationship exist

between the two.

Choi and Kim (2000) empirically examined several major determinants

of American Depository Receipts (ADRs) and their underlying stock returns.

The end-of-week closing American Depository Receipts prices for the period

of 1990-1996 were obtained from the Centre for Research in Security Prices

(CRSP) database. To get the Morgan Stanley Capital International Perspective

Index, exchange rates and underlying stock prices, the Datastream database

also has been employed. In order to examine the behaviour of American

Depository Receipts and the underlying stock returns overtime, the total

period was divided into two subperiods. They are: (i) 1990-1993, and (ii)

1994-1996. The effect of changes in exchange rate on American Depository

Receipts returns was examined using the regression technique.

The analysis reveals that the behaviour of the American Depository

Receipts can be explained better using the exchange rates. By and large,

exchange rate shows a significant negative correlation with American

Depository Receipts returns. The authors concluded by observing that the

United States investors are discouraged about investing in American

Depository Receipts, when the dollar is strong. This is due to the fact that

weak foreign local markets with the strong dollar may adversely affect local

firms issuing American Depository Receipts. The strong dollar reflects a

sound United States economy, and United States investors have less incentive

to invest in American Depository Receipts, especially when the value of

foreign firms denominated in the United States dollar declines.

Gao (2000) attempted to examine the effects of unexpected exchange

rate movements on the stock returns of United States multinationals. For the

empirical analysis monthly returns on individual stocks traded in the New

York Stock Exchange or American Stock Exchange over the period from

111988 to 911993 were selected from the Centre for Research in Security

Prices (CRSP) 1993 Master File. Monthly Standard and Poors 500 stock index

returns were used for the market portfolio returns. Exchange rates between the

dollar and foreign currencies came ffom the National Trade Data Bank. For

each industry, the exchange rate used in the estimations is a detrended log

export weighted exchange rate. The or- least square method was adopted

for estimation purpose. Levinsohn and Mackne - Mason's two stage

estimation procedure was also employed because of the biased and

inconsistent estimates of the one-step method.

The empirical study suggests that the stock market correctly reveals the

profitability effects of unanticipated exchange rate changes. A depreciation

shock of the United States dollar has significant positive effects on the

abnormal retwns on the stocks of the multinationals where as, an appreciation

causes a negative returns of the stocks.

lorio and Faff (2000) enquired the foreign exchange exposure of the

Australian equities market, being one of the most established markets in the

Asia-Pacific. The data employed were continuously compounded daily and

monthly returns on 24 Australian industry indices, and it was obtained from

Datastream database. The period of the analysis involves 2280 (108) daily

(monthly) observations from January 1988 to December 1996. The exchange

rate factor's return was based on Australian dollarNnited States dollar

exchange rate. For the empirical analysis, an augmented market model which

consists of the return on the market index and the return on the exchange rate

factor as independent variable has been tested.

The study reveals that while employing daily data, there is some

evidence of significant exchange rate exposure of the predicted signs in nine

industries. As far as the monthly data is concerned, results are found less

encouraging.

Kearney (2000) enquires how volatility is transmitted to national stock

markets from their international counterparts as well as from domestic

business - cycle variables. The author has used low - frequency data

consisting of end monthly observations on stock market indices for five

counmes (Britain, France, Germany, Japan and the United States) over the

period from July 1973 to December 1994. The stock market indices employed

in the study are The Financial Times All Share Index (FTSE), the Paris

General index, the Frankfiu-t index, the Nikkes 225 index and the Dow Jones

30 composite index. The indices are converted into United States dollar at the

prevailing end of month spot bilateral exchange rates taken on the last stock

market trading day. All data have been extracted from the Datastream

International Database. The time series propemes of the data were examined

by employing Dickey - Fuller and Phillips - P m n tests. The Johansen

multivariate cointegration technique and error correction model (ECM) was

considered for examining the nexus between the two. Besides, the volatility

transfer between the World, European, and J a p d n i t e d States stock markets

was estimated by employing the generalised least squares (GLS) method.

The study reveals that the national stock market indices are multivariate

cointegrated with their domestic business cycle variables. Britain is found

more volatile to the exchange rate and United States has the lowest volatility.

Penttinen (2000) analysed the devaluation risk related peso problem in

stock retums. The author aims at providing a rational explanation for the

longer non-random negative trend in the Finnish stock market in the 1989-

1992 period. For the empirical analysis, daily data were collected from the

Helsinki stock exchange. Data on Finnish Markka came from the Bank of

Finland. The author has used cross-sectional regression analysis on the

individual firm level.

The analysis reveals that the seemingly anomalous negative trend in the

Finnish stock market from June 1989 to the devaluation of the Finnish markka

(FIM) in November 1991. It was partially caused by devaluation-risk-related

peso problem in individual stocks.

Tai (2000) investigated the role of exchange rate risk in pricing a

sample of the United States commercial bank stocks by estimating and testing

a three-factors model under both unconditional and conditional framework.

The author has taken weekly sample of 31 commercial bank stock, traded on

the New York and American stock exchanges. Then it was divided into three

groups namely, the money Center bank, the large bank and the regional bank.

The world market risk was measured as the United States dollar return of the

Morgan Stanley Capital International (MSCI) world equity market in excess of

7-day eurodollar deposit rate the exchange rate risk was measured as the log

first difference in the trade - weighted United States dollar price of the

currencies of 10 industrialised countries. All the data were obtained from

DATASTREAM. The empirical estimation and testing were conducted by

using three econometric methodologies, namely, Nonlinear Seemingly

unrelated Regression (NLSUR) via, Hansen's Generalised Method of Moment

(GMM), Pricing Kernel approach and the multivariate Generalised

Autoregressive Conditional Heteroskedasticity in mean (MGARCH-M)

model. The pricing kernel approach reveals a strong evidence of exchange rate

risk in both large bank and regional bank stocks. For the regional banks world

market risk was also found positive. Finally, estimation based on the

multivariate Generalised Autoregressive Conditional Heteroskedasticity in

mean (MGARCH-M) approach shows strong evidence of time - varying

exchange rate risk prernia and weak evidence of time varying market risk

premiam for all three bank portfolios.

Koch and Saporoschenko (2001) examined the sensitivity of individual

and portfolio stock returns for Japanese horizontal Keiretsu financial firms to

unanticipated changes in market returns, bond returns, exchange rate changes

and nominal interest rate spread changes. For the empirical analysis, weekly

stock returns from January 14, 1986 through December 29, 1992 were

collected. Weekly traded weighted yen exchange rate return innovation was

estimated using data from the JP Morgan economic department for the same

period. Generalised Autoregressive Conditional Heteroskedasticity (GARCH)

model has been employed to examine the stock return sensitivity of Japanese

horizontal Keiretsu financial firms to exchange rates. The results indicate that

Keiretsn financial fums have insignificant exposure to exchange rate changes.

Nagayasu (2001) analysed empirically the Asian fmancial crisis by

using high - frequency data of exchange rates and stock indices of the

Phillippines and Thailand. Daily data on benchmark stock indices and

exchange rates covering the period from 11-15-1996 to 12-31-1998 were

obtained from the Bloombery dataset. To analyse the relationship between

exchange rates and stock indices, the author focused on causality among these

variables using the method developed by Granger (1969). The robustness of

findings was examined using the threshold auto-regression. The Granger non

causality test in two subsets of data has been conducted in order to take into

account a possible regime shift.

The analysis shows the importance of the banking and financial sector

in explaining the Asian crisis. The result from the threshold method provide

evidence for a negative relationship between the exchange rate and the stock

indices i.e. a fall in stock prices is associated with currency depreciation. This

paper also identifies contagion effect running from Thailand to Philippines.

Nieh and Lee (2001) explored the dynamic relationship between the

stock prices and the exchange rates, for each G-7 countries. The data were

collected from Dow Jones News1 Retrieval provided by Dow Jones, Inc, for

the sample period from October 1, 1993 to February IS, 1996, of daily closing

stock market indices and foreign exchange rates. To test stationarity of the

data authors have employed the Augmented Dicky Fuller and the multiple -

unit - root test suggested by Dickely and Pantula respectively. They adopted

the cointegration test to avoid the spurious regression problem. For which

Engle and Granger two - step methodology and Johansen Multivariate

Maximum Likelyhood cointegration test were also conducted. To capture both

the short - run dynamic and the long-run equilibrium relationship Vectors

Error Correction model (VECM) was also employed.

The findings reject any long-run significant relation between stock

prices and exchange rates. But one day's significant shorterm relationship has

been found in certain G-7 countries. Analysis also reveals that the difference

in the results among G-7 countries might be due to the influence of many other

country specific factors in addition to the two financial assets.

Sadorsky (2001) studied the stock returns of Canadian oil and gas

industry using a multi-factors market model which allows for several risk

premiums including that of the exchange rates. The data collected for the

study are on monthly basis and they cover the period From April 1983 to April

1999. All economic data were obtained from the Statistics Canada economic

database. Oil price shares are measured using the Toronto Stock Exchange oil

and gas index. The monthly United States dollar i Canada dollar exchange

rates also have been collected for the analysis. Both the Dickey Fuller (1979)

and the Phillips and Person (1988) tests were employed to test the stationarity

of the data. Standard error for the parameter estimates were computed using

the Newey and West heteroskedasticity and autocolrelation coefficient

consistent covariance matrix.

The author indicates that oil and gas stock returns are sensitive to

several risk factors including the exchange rate. It has also been found that an

increase in exchange rate decreases the return to Canadian oil and gas stock

prices. Hedging exchange rate risk will allow the oil and gas industry more

flexibility through better management of cash flow.

Wu (2001) investigated macroeconomic exposure including the

exchange rate, of a major Asian Stock index namely, the Straits Times

Industrial Index (STII) of Singapore. The study was conducted for both the

1997-98 Asian financial crisis and the pre-crisis periods. The authors

employed monetary approach to analyze the asymmetric asset-price

movements (exchange rate and stock prices) in Singapore, a small open

economy with managed exchange rate targeting. The Primask Datastream

provided the data on the bilateral Singapore dollar exchange rate vis-a-vis its

major trade partner's currencies, with the data ranging from January 22, 1990

to December 12, 1997. The distributed lag model and the Vector

~utoregression (VAR) approach have been employed to analyze the

macroeconomic exposure.

Analysis reveals that monetary policy does not play a significant role in

such a small economy with exchange rate targeting. But fiscal revenue as well

as fiscal expenditure exerts positive influence on the Straits Times Industrial

Index (STII). It is also found that the Singapore dollar exchange rates vis-a-vis

the developed countries' currencies are negatively related with the Straits

Times Industrial Index (STII) both before and during the 1997-1998 crisis

period, whereas the exchange rate in relation to the Malaysian ringgit is

positively related to the Straits Times Industrial Index (STII).

Baharumshah et a1 (2002) tested an augmented monetary model that

includes the effect of stock prices on the bilateral exchange rates. The model

was employed to the ringgit 1 United States dollar (RM/USD) and ringgit 1

Japanese yen (RMIJY) exchange rate. The quarterly data on exchange rates as

well as the macroeconomic variables were collected for a period from the first

quarter of 1976 to the last quarter of 1996. The stock market is represented by

the main stock index. The exchange rates were obtained h m the International

Monetary Fund's International Financial Statistics (IFS) and Malaysia is

regarded as the home country. Stock indices were from Morgan Stanley

Capital International. All the variables apart from interest rates are expressed

in logarithmic differentials. The standard Augmented Dickey - Fuller and

Phillips - Pelron tests were employed to test for the order of integration of all

the series. The Johansen (1988) maximum likelihood technique was also used

to test for long-run cointegration.

The empirical fmdings reveal that the equity market is significant in

affecting the exchange rate and in explaining at least in part the parameter

instability evidenced in the cointegrating system. The authors conclude that

models of equilibrium exchange rate should be extended to include equity

markets, in addition to bond markets.

Chang (2002) attempted to investigate industry - level currency risk of

Taiwan's stock market around the Asian financial crisis. By dollar value of

transaction, the Taiwan stock market is ranked third in the world. The data

used in this study consists of daily New Taiwan Dollar (NTD)Nnited States

Dollar (USD) bilateral exchange rates, trade weighted effective New Taiwan

Dollar (NTD) rates and industry level stock indexes h m 1 January, 1996 to

31 October, 1998. The bilateral New Taiwan DollarNnited States Dollar

(NTDNSD) exchange rates quoted by the Bank of Taiwan were retrieved

from the Taiwan Economic Journal. The trade weighted effective New Taiwan

Dollar, calculated from the exchange rates of 17 countries was considered

from the Council for Economic Planning and Development. The daily stock

returns retrieved from the Taiwan Economic J o d consist of f m listed in

the Taiwan Stock Exchange (TSE) and over-the counter (OTC) securities

exchange. The impact of exchange rate exposure on index returns was

discussed using a Generalised Autoregressive Conditional Heteroskedasticity

(GARCH) regression framework.

The analysis reveals that over 50 per cent of industries have statistically

significant currency exposure over the entire sample period, when bilateral

New Taiwan Dollar (NTD)Nnited States Dollar (USD) exchange rates were

used as currency risk factors. The author concludes by pointing out that

exchange rate risk is less for larger firms than for smaller firms.

Chen and So (2002) examined the effect of Asian financial crisis on the

sensitivity of United States multinational f m s to United States stock market.

The companies have been divided into sample group and control group. The

sample group was comprised of multinational firms with sales in Asia -

Pacific region and control group was comprised of multinational firms with

overseas sales outside Asia-Pacific region. The study period was from January

1996 to December 1998. This was again divided into the subperiod of 1%

years before and afier crisis. Weekly data on exchange rate between the

United States dollar and foreign currencies have been collected from the

DATASTREAM. The annual measures of foreign sales and foreign assets for

each sampled company were collected from the COMPUSTAT data base.

Weekly return on individual stocks of the firm in the sample group and the

market portfolio were also obtained from the DATASTREAM. Finns in the

control group also came from the COMPUSTAT data base. A continuous

weekly stock return index of control f m were also collected from the

DATASTREM. The authors tested the difference in stock return variance

between the two sub-periods, by employing the X2 statistic. They also

performed a non parametric Wilcoxon Signed - rank test and compared the

median of stock return variance across the two sub-periods. The model

employed for the analysis was based on the Capital Asset Pricing Model

(CAPM).

The study found that the variance of stock returns for the sample firms

increased dramatically during the second subperiod when the exchange rates

of Asian currencies were unusually volatile. The variance of the control firms

was not dramatic as that of the sample firms. It again reveals that United

States multinational firms with sales in Asia - Pacific region showed a

significant increase in market risk corresponding to the increase in exchange

rate variability across the two sub-periods before and after the Asian financial

crisis.

Iorio and Faff (2002) examined the pricing of foreign exchange risk in

the Australian equities market. The period of study was from 1 January 1988

to 30 September 1998. The authors have employed both daily and monthly

returns data. For the daily analysis 2723 observations were used. whereas 128

observations were analysed for the monthly returns. All data were obtained

from DATASTREAM. To test the stationarity of the data employed,

Augmented Dickey Fuller (ADF) testing procedure was applied. Phillips -

Perron (PP) test was also used to test the robustness of the results. In addition

to this a basis two factor 'market and exchange rate' asset pricing model, a

'zero-beta' version of the two-factor model and an orthogonalised two-factor

model were also employed to examine the objective.

The analysis shows that foreign exchange risk is priced for the full

sample period 1988-1998. It is also revealed that foreign exchange risk is

more prominent in the two sub periods viz; 1990-1 993 and 1997-1 998. In

these two periods both the Australian economy and the Australian dollar were

relatively weak and uncertain.

Patro et a1 (2002) studied the significance as well as the determinants of

foreign exchange risk exposure for equity index returns of 16 Organisation for

Economic Co-operation and Development (OECD) countries. Weekly

observations of equity index prices were obtained from Morgan Stanley

Capital International (MSCI) for 16 Economic Co-operation and Development

(OECD) countries. Sample on returns starts in January 1980 and covers up to

December 1997 for a period of 18 years. Weekly dollar exchange rates for

these countries were obtained from the Federal Reserve Board for the same

period. A generalised autoregressive conditional heteroskedasticity (GARCH)

specification was used to find the significant time - varying foreign exchange

risk exposure. A generalised least square (GLS) regression has been employed

to examine the impact of exchange rate betas on specific macro economic

variables. To improve estimation efficiency, data for all 16 countries were

pooled and ran a panel regression.

The analysis reveals a significant time-varying currency betas for

country equity index returns. It has also been found that the exchange rate risk

exposures are related to a country's macroeconomic aggregates.

Wongbanpo and Sharma (2002) investigated the interdependence

between stock market, and fundamental macroeconomic factors including the

exchange rate, in Association of South East Asian Nations-5 (ASEAN-5)

countries i.e. Indonesia, Malaysia, Philippines, Singapore and Thailand. The

authors have collected monthly data from 1985 to 1996 from the World Stock

Exchange Fact Book, Datastream, and the June 1999 volume of International

Financial Statistics. All series were transformed into natural logs prior to the

empirical analysis. To test the stationarity of each of the series Dickey -

Fuller, Augmented Dickey - Fuller (ADF) and Philips and Pewon tests were

employed. Besides, they also employed the maximum likelihood based hmax

and ktrace statistics introduced by Johansen (1988, 1991) and Johansen and

Juselius (1990) to test the number of significant cointegrating vectors. The

likelihood tests were done to determine the lag length of the vector auto

regressive system. The study found that the effect of exchange rate is positive

in Indonesia, Malaysia and Philippines, and it is negative in Singapore and

Thailand.

Abid et a1 (2003) tested for evidence of contagion between the stock

markets of eleven Asian countries as an important cause for the spread of the

Asian crisis. They also examined cross - country co-movement among the

rates of returns of cmency markets and stock markets. Weekly data on stmk

returns of United States and eleven Asian stock markets were drawn kom the

~nternational Financial Corporation (IFC) database published in the 'Financial

Times' for the period from 1 September 1989 to 29 October 1999. Data on

weekly foreign exchange rate of returns for Asian currencies against the

United States dollar were available on the OANDA database, published on

internet, for the period from January 1, 1990 to October 29, 1999. An

empirical analysis has been carried for the entire period and for the two sub-

periods i.e. the sub-period before the Asian crisis and the sub-period during

and after the Asian crisis, by using the univariate Generalised Autoregressive

Conditional Heteroskedasticity (1, 1) model to study the conditional volatility

time varying in financial and foreign exchange market.

The authors fmd evidence of an increased level of co-movements in

cross-counby excess return in the sub-period that encompassed and followed

the Asian crisis as compared to before the crisis. The findings also underline

the presence of cross-border contagion in both currency and equity market.

Bailey et a1 (2003) tested the impact of switching between silver, gold

and paper money standards on stock returns from seven small open

economics. The sample was collected for a period from December 1872 to

November 1941. End of month stock prices were collected ftom principal

national, colonial or metropolitan news papers. The authors have translated all

stock prices into pounds using end-of-month exchange rate, and then its log

differences were also computed. To perform three diagnostic tests Engle and

Ng Generalised Autoregressive Conditional Heteroskedasticity (GARCH)

specification was used.

The analysis reveals that the paper money regimes are more likely to be

associated with heightened stock market volatility and correlation between

stock indexes. Under the traditional silver regime stock markets did not draw

closer. It is also found that correlations between index returns and China's

exchange rate weaken when China was on paper regime. This shows that

conditions and events beyond the currency system affect the variances and

correlations.

Bin et a1 (2003) considered the role of exchange rate as a pricing factor

for American Depository Receipts (ADRs). The sample of American

Depository Receipt consists of 125 issues grouped into 11 country specific

portfolios were obtained from the Internet Search engine provided on the Bank

of New York Depository Receipts Directory Website. Weekly returns of each

sample American Depository Receipts were collected from the centre for

Research in Security Prices (CRSP) data tape covering the span between 1

January 1990 and 31 December 2000. Weekly observations of foreign equity

market indices, bilateral spot exchange rates and United Stateslforeign money

market rates came from Datastream International. In order to value American

Depository Receipts portfolio, a modified version of the Arbitrage Pricing

Theory (APT) model formulated by Ross (1976) has been adopted. For the

purpose of estimation the authors employed both Generalised Autoregressive

Conditional Heteroskedasticity (GARCH) and Seemingly Unrelated

Regression (SUR) models.

The study reveals the importance of foreign exchange rates as a

determining factor of American Depository Receipt pricing. Therefore, the

exchange rate risk premium matters in investment decisions involving

American Depository Receipts. It is also found that the outbreak of a foreign

financial crisis has a negative effect on the pricing behaviour of American

Depository Receipts originating in the country or region.

Johansen and Soenen (2003) investigated the factors that affect the

level of economic integration and stock market co movement between the

United States stock market and eight American equity markets, namely, of

Argentina, Brazil, Chile, Mexico, Canada, Colombia, Peru and Venezuela. For

the empirical analysis, Morgan Stanley Capital International's daily closing

stock index values were collected for all nine national equity markets. Data

were taken for the period 1988-1999, except for Colombia, Peru and

Venezuela, for which it was only available from 1993 to 1999. All other

necessary information including that of the exchange rate was obtained from

Datastream. The Geweke measures of feedback were estimated to analyse the

significance of factors affecting the level of economic integration and stock

market co movement.

The authors observe that a higher share of international trade in terms

of exports and imports with the United States have a strong positive effect on

stock market co-movements between country pairs. Increased bilateral

exchange rate volatility in the United States relative to other countries

contributed to lower co movement.

Kiyrnaz (2003) investigated the foreign exchange exposure of firms

traded on the Istanbul Stock Exchange (ISE). The study period was from

January 1991 to December 1998. It was further divided into the pre-crisis

period from January 1991 to February 1994 and the post - crisis period from

May 1994 to December 1998. The author has taken a sample of 109 firms

from the Istanbul Stock Exchange for analysis. Based on the firms' industry

affiliation sub-samples were also constructed namely, chemical/petroleum,

food/beverage, machinery/ equipment, basic metal, non-metall cement, wood/

Paper products, textiles, financials and service 1 trade. The sample was again

classified based on the foreign involvement of the firms. To measure the

foreign exposure of firms a foreign exchange basket composed of 1 United

States Dollar and 0.77 European Currency Unit was constructed. The

exchange rate exposure has been estimated using a time series regression.

The analysis reveals that the firms are highly exposed to foreign

exchange risks. The level of exposure is more pronounced for the textile,

machinery I equipment, chemical / petroleum and financial industries and less

in the foodbeverage, service1 trade and non-metal / cement industries. It is

also found that the post-crisis exposure tends to be lower than that of the p re

crisis period. This shows more attention given by the fums to the foreign

exchange exposure after the crisis.

Lin et al (2003) scrutinized the hypothesis that the euro has become a

major influence on international stock markets. All data used in the study were

weekly data between 1 January 1999 and 31 December 2002 and they are

collected from datastream. The Financial Times Stock Exchange (FTSE) 100

Index was used as a proxy for the London Stock Market. A technique of zero-

non-zero (ZNZ) patterned vector auto regressive modelling was employed to

examine the causal relation between the euro and the London Stock market

The findings indicate that movements in the euro are related to

movements in the London market. It is also observed that both a current shock

and a lagged shock to the euro's forex market impacts on the movements of

both local London stock market and forex market.

Shamsuddin and Kim (2003) attempted to study the extent of stock

market integration between Australia and its two leading trading partners, the

United States and Japan. In addition, this study determines whether the extent

and nature of stock market integration in the period of the post - Asian crisis

differs from that of the pre-Asian crisis. The data used in this study are the end

- of week closing stock price indexes for Australia, Japan and the United

States, and the Australian dollar value of the Japanese yen and United States

dollar. The national stock indexes used were the Standard and Poors 500

composite index for the United States, the Tokyo Stock Price Index (TOPIX)

for Japan and the All Ordinaries Index (AOI) for Australia. The pre-Asian

crisis period covers two sub-periods i.e. from January 1991 to December 1993

and from January 1994 to July 1997. The post-Asian crisis period covers from

January 1998 to May 2001. All data were collected from Data stream.

Augmented Dickey Fuller (ADF) test for unit root was adopted to test the

stationary property of each variable. Co-integration technique was employed

to examine long-run co-movement of stock prices for Australia, Japan and the

US, and the Australian dollar value of the Japanese yen and US dollar. To

understand the dynamic linkages among national stock prices as well as the

interaction between stock prices and exchange rates, a Vector Error Correction

Model (VECM) was employed for two sub periods ,and Vector

Autoregressive (VAR) model in first difference was employed for the post-

Asian crisis period.

The authors concluded that there was a stable long-run relationship

among the Australian, United States and Japanese market prior to the Asian

crisis, but this relationship disappear during the post-Asian crisis period. It is

also found that following the Asian crisis, the influence of United States on the

Australian market diminished while the influence of Japan remained at a

modest level.

Bartram (2004) investigated the linear and nonlinear foreign exchange

rate exposure of German nonfinancial corporations. For the purpose of

analysis the data on stock prices came from the Datastream International. The

exchange rates French Franc (FRF), Dutch Guilder (NLG), Italian Lira (ITL),

British Pound (GBP), United States Dollar (USD), Belgian Franc (BEF),

Swiss Franc (CHF), Australian schilling (ATS) and Japanese Yen (JPY) were

available from the Deutsche Bundesbank. For the empirical examination a

regression model was estimated.

The empirical evidence does not indicate that the economic foreign

exchange rate exposure is primarily driven by the currency of determination.

Firms with more international sales exhibit systematically larger and more

significant foreign exchange rate exposures.

Chen et a1 (2004) investigated the firm value sensitivity to exchange

rate fluctuation by focussing mainly on individual fums and also looked at the

differing rate of sensitivity between currencies. For the empirical analysis a

sample of 161 New Zealand Stock Exchange (NZSE) listed firms were

considered. Monthly share return indexes were obtained from the Global

Datastream database for the period from January 1993 to December 2000. The

New Zealand (NZ) dollar, trade - weighted Index (TWI), the exchange rates

for US dollar and Australian dollar were obtained from the web page of

Reserve Bank of New Zealand. The trade - weighted Index (TWI) is

calculated using the rates of the five currencies of New Zealand's five main

trading partners (Australian Dollar - 38 per cent, Japanese Yen - 24 per cent,

United States Dollar - 22 per cent, Great Britain Pound Sterling - 10 per cent

and Euro - 6 per cent). Test was conducted using a residual regression model.

The cross sectional analysis was done by employing the multivariate

regression model.

The estimation results reveal strong evidence that exchange rate

movements affect the value of the listed New Zealand firms. It is also

observed that firms are on average positively related to the movement of the

United States dollar and negatively related to the movement of the Australian

dollar i.e. firms gain in value when the New Zealand dollar appreciates against

the United States dollar and depreciates against the Australian dollar.

Grammig et al (2004) examined exchange rates along with equity

quote's for 3 German firms from New York (NYSE) and Frankfurt (XETRA)

during overlapping trading hours to know where price discovery occurs and

how stock prices adjust to an exchange rate shock. The three German firms

considered for the study are Daimler - Chrysler (DCX), Deutsche Telecom

(DT) and SAP. Data were taken for the period from August 1 to October 3 1,

1999. The New York Stock Exchange (NYSE) quotes data were taken from

the TAQ set available from the New York Stock Exchange. The XETRA

quote data came from the Frankfurt Stock Exchange. Information on the

exchange rates is tick - by tick quotes on the dollar prices of the euro as

reported on the Reuters indicative quoting screen. This was obtained from

Osten and Associates, Zurich. Augmented Dickey - Fuller test was conducted

for stationarity. Johansen cointegration tests were performed for the presence

of cointegrating vectors among the variables. The Schwan Information

Criterion (SIC) has been employed to identify the appropriate lag length. The

estimation precision was also assessed by employing the bootstrap method

suggested by Li and Maddala.

The study shows that price discovery occurs largely in the home

market. It is also revealed that exchange rate innovations have a large impact

on the New York (NYSE) price but a very small effect on the Frankfurt

(XETRA) price.

Priestley and Odegaard (2004) explored the pricing of exchange rate

risk while simultaneously analysing the effect of long swings in dollar on the

price of exchange rate risk. The authors used both the dollar Yen (JPY) and

dollar European Currency Unit (ECU) rates because of their large weights in

United States imports and exports. To capture the source of systematic risk in

the economy excess return on the aggregate United States stock market

portfolio and four macroeconomic factors such as the changes in industrial

production, the change in inflation, the term spread and the default spread

were also included. The monthly data on stock market return were collected

from Kenneth French. The data on macroeconomic factors were obtained from

the Federal Reserves. The period of study was from 1979 to 1990. To jointly

estimate the parameters of the asset pricing model a Nonlinear Seemingly

Unrelated Regression (NLSUR) framework was used.

The study found that the Yen and European c ~ e n c y unit are both

priced and the prices of risk are different under different regimes. This is due

to the extent of export and import to the European Union and Japan. The

analysis again reveals the importance of the changes in exchange rate regimes

for the f m s . When the dollar appreciates, investors would b a n d a premium

to hold stocks of companies that are exporters. At the same time investors in

importing f m would be willing to pay premium to hold the stocks. This is

important from the point of view of risk management.

National-level Studies

Apte (1998) addressed the question of whether stock markets face

exchange rate risk. For the analysis main source of data was the Centre for

Monitoring Indian Economy (CMIE) corporate database which provided data

on individual stock prices as well as on the Sensitive Index (SENSEX). For

the exchange rate factor, data on nominal and real effective exchange rates

were obtained from the Reserve Bank of India bulletin. The period of study

was from May 1990 to May 1997. The empirical test was carried out using a

two factor model, for which an ordinary least squares (OLS) regression

techruque was employed.

The study reveals the presence of exchange rate risk as a systematic

risk factor over and above market risk for a number of Indian companies. The

main determinant of a firm's exchange rate exposure appears to be the

importance of exports in its total turnover. Import intensity of its operation

also shows some influence but it is not statistically significant.

Jithendranathan et a1 (2000) attempted to evaluate the market

segmentation and its effect on the pricing of cross listed securities using Indian

Global Depository Receipts (GDRs). For the purpose of analysis the Skindia

Global Depository Receipts index as a market index proxy for the

performance of the Indian Global Depository Receipts, was collected for the

period from 1995 to 1998. This index consists of 18 actively traded Global

Depository Receipts. Monthly Global Depository Receipts dollar prices and

the premiums on the same were obtained from Skindra Finance Private, India.

Monthly Standard and Poors 500 index returns and Financial Times 100 Index

returns have also been used from the index values provided by Commodity

Systems. Both single index models and multi-factor models have been adopted

to test the effect of both domestic and foreign factors on the Global Depository

Receipts returns and the underlying security returns.

The authors observed that Global Depository Receipts index returns are

affected by both foreign and domestic factors, while the underlying Indian

securities are influenced only by domestic factors. It is also found that Global

Depository Receipts are priced at a premium over the exchange rate adjusted

prices of the underlying Indian securities.

Damele et al (2004) attempted to analyse the market integration based

on the stock market and the exchange market. For the analytical purpose the

indices such as Bombay Stock Exchange Sensex, Bombay Stock Exchange

National and Nifiy have been adopted as the representative of Capital market.

The Rupee Dollar exchange rate was considered to see the movement in Forex

market. All these data were collected from the various issues of Economic

Times. The period of study was form 1991 to 2002. The empirical analysis

was carried out by using the techniques such as the coefticient of variance and

the regression analysis.

The empirical results about the price integration between stock prices

and exchange rates lead to the conclusion that there exists some degree of

price integration between these markets is India.

Conclusion

From the review of relevant literature, it is observed that for the last one

and a half decade, a number of studies have taken place to examine the

relationship between foreign exchange market and the stock market. Based on

the classification of empirical studies discussed earlier, it is found that most of

the studies are at the international level. In the international context, authors

such as Mao and Kao (1990), Jorion (1991) Bartov and Bodnar (1994), Choi

and Prasad (1995), Prasad and Rajan (1995), Bartov et. a1 (1996), Charnberlin

et. a1 (1997), Mookcrjee and Yu (1997), Antonious et. a1 (1998),

Malliaropulos (1998), Bodart and Reding (1999), Ding et. al (1999), Kwon

and Shin (1999), Choi and Kim (2000), Iorio and Faff (2000), Koch and

Saporoschenko (2001), Bailey et. a1 (2003), Chen et. al(2004) etc. empirically

examined the inter-linkages between the foreign exchange market and the

stock market. Most of the studies show mixed results. Some studies have

found a significant positive relationship between stock prices and exchange

rates, while others have reported a significant negative relationship between

the variables.

Some of the studies like Solnik (1987) and Smith (1992) have reported

a significant positive relationship between the stock prices and the exchange

rates, while other studies such as Eun and Rensik (1988) have found a

significant negative relationship. At the same time, studies like Franck and

Young (1972) have revealed very weak association between stock prices and

the exchange rates.

In addition to the above studies, some other studies examined the

causation between the stock prices and the exchange rates. When studies such

as Morley and Pentecost (1998) have reported causation from exchange rates

to stock prices, other studies have found causation from stock prices to

exchange rates [eg: Ajayi and Mougoue (1996)J. On the other hand, Bahmani-

Oskooee and Sohrabian (1992) claimed bidirectional causation between stock

prices and exchange rates in the short run.

The earlier literature again reveals that majority of the studies are in the

context of developed economies. Not only that many of the empirical studies

are giving different and conflicting results also. Some studies found positive

effect of foreign exchange rate on stock prices, while some others are showing

negative impact on it. It is also found that most of the studies have adopted

general stock market indexes for the analysis, instead of analysing the case of

firms or industries separately.

From the methodological point of view, the majority of the studies have

used one directional analysis. For this purpose authors have employed

econometric models such as ordinary least squares (OLS) and generalized

least squares (GLS). Only very few studies have examined the market

integration with reference to stock market and foreign exchange market.

Besides, much emphasis has not been given in the literature to examine the

inter-relationship between stock market and foreign exchange market. Not

only that, for the above purposes most of the authors have adopted methods

such as Granger Causality test, and it assumes that time series are stationary in

level, which can be highly restrictive for stock price or exchange rate time

series. But, the techniques such as Johansen's co-integration, Vector

Autoregressive model and Vector Error Correction models are underlining the

problem of non-stationary of the data with unit roots. The present study is an

earnest attempt in this direction.

Another important lacuna of the existing literature is that in the case of

India very few studies have taken place. The liberalisation and the

globalisation of 1990s in lndia again show the importance of a re-examination

of the linkages between foreign exchange market and the stock market.

Liberalized exchange rate system, opening up of the capital account for

international investment, the advent of floating exchange rates, the

development of 24-hour screen based global trading, the increased use of

national currency outside the country, innovation in internationally traded

financial products, recent spurt in the Foreign Institutional Investments (FII),

and the introduction of American Depository Receipts (ADR) and Global

Depositors Receipts (GDRs) after 1992, are multiplying the relevance of .a

detailed analysis in the above mentioned direction in Indian context.

On the basis of the above observations, the present study attempts to

examine the validity of market integration with reference to foreign exchange

market and stock market and its inter-relationship in India. For a clear

understanding of the interrelationship between the exchange rates and the

stock prices study has been carried out both at the firms and at the industry

levels. In addition to this, for a meaningful comparison analysis has also been

carried out by employing the market index (BSE Sensex).